Acute kidney injury (previously known as acute renal failure) has a high morbidity and mortality. After developing a novel model of sepsis-induced AKI that employs cecal ligation puncture in elderly mice treated with fluids and antibiotics, we are using the model to study the pathophysiology of injury, to screen drugs, and to study their mechanisms of action, including conscious blood pressure by telemetry. We adjusted our mouse model by using outbred mice, which develop AKI at a younger age, and we established another model using comorbidity, namely pre-existing renal dysfunction, which is thought to increase susceptibility to AKI in patients. We continue to study both the sepsis-AKI model and the 'acute-on-chronic'sepsis-AKI model. 1) Following up our studies on ethyl pyruvate, we used a more stable analog, methyl-2-amidoacrylate (M2AA), and we found that M2AA could benefit mice up to 6 hours after induction of sepsis. Although NFkappaB was transiently increased in spleen, peaking at 6 hours, and NFkappaB is a known target of M2AA in vitro, removal of spleen did not decrease the overall effectiveness of M2AA. However, because we could eliminate the spleen as the target of M2AA, and we established that the spleen is required for chloroquine effectiveness, we used M2AA in combination with chloroquine with some success. 2) We collaborated with Eva Mezey and Krisztian Nemeth of NIDCR to test the effectiveness of bone marrow stromal cells (BMSC, also known as mesenchymal stem cells) on sepsis. We found that BMSC were effective in suppressing organ damage and mortality from sepsis, and examined the mechanism. First, we found that BMSCs were short-lived and targeted primarily to the lung, implicating a paracrine effect. The BMSCs were adjacent to macrophages in lung, and in a series of mechanistic experiments we determined that macrophages were essential for the benefit of BMSCs. In similar experiments we determined that IL-10 was essential for beneficial effects of BMSCs, and that BMSCs increased IL-10 production by activated macrophages. BMSCs required intact Toll-like Receptor 4/MyD88 signaling, as well as cyclo-oxygenase 2 activation and PGE2 production;macrophages required both EP2 and EP4 to detect PGE2. Therefore, we concluded that activated macrophages appear to trigger BMSCs to secrete PGE2 locally to downregulate the macrophage response.